Efficient Electrocardiogram-based Arrhythmia Detection Utilizing R-peaks and Machine Learning

Pham, Van Thinh; Pham, Van-Su; Nguyen, Minh Tuan; Le, Hai-Chau

doi:10.1109/icsse58758.2023.10227145

Cited by 1 publication

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, all these AutoML strategies require model training to estimate the quality of the designed networks [16], which can compromise the efficiency and accuracy of the model search [17, 18]. DNN training is too time-consuming and computationally costly, routinely requiring hundreds [19] to dozens of thousands [20] of GPU hours; thus, the use of AutoML methods for clinical DNNs in the broader clinical and biomedical community has been prohibitive.…”

Section: Mainmentioning

confidence: 99%

Training-free Design of Deep Networks as Ensembles of Clinical Experts

Wu,

Chen,

Zhang

2024

Preprint

View full text Add to dashboard Cite

Artificial intelligence (AI) techniques such as deep learning hold tremendous potential for improving clinical practice. However, clinical data complexity and the need for extensive specialized knowledge represent major challenges in the current, human-driven model design. Moreover, as human interpretation of a clinical problem is inherently encoded in the model, the conventional single model paradigm is subjective and cannot fully capture the prediction uncertainty. Here, we present a fast and accurate framework for automated clinical deep learning, TEACUP (training-free assembly as clinical uncertainty predictor). The core of TEACUP is a newly developed metric that faithfully characterizes the quality of deep networks without incurring any cost for training of these networks. When compared to conventional, training-based approaches, TEACUP reduces computation costs by more than 90% while achieving improved performance across distinct clinical tasks. This efficiency allows TEACUP to create ensembles of expert AI models, mimicking the recommended clinical practice of using multiple human experts when interpreting medical data. By combining multiple perspectives, TEACUP provides more robust predictions and uncertainty quantification, paving the way for more reliable clinical AI.

show abstract